Germline variants and their role in the pathogenesis of immune effector cell-associated hemophagocytic lymphohistiocytosis-like syndrome (IEC-HS) in adult lymphoma patients

Abstract

Abstract Introduction: IEC-HS is a rare but potentially fatal adverse event following chimeric antigen receptor (CAR) T cell and other IEC-engaging therapies. Clinically, IEC-HS presents as a severe hyperinflammatory syndrome, resembling primary or familial hemophagocytic lymphohistiocytosis (HLH), which typically manifests in childhood, as well as secondary HLH triggered by infections, malignancy, or autoimmune disorders. While primary HLH is caused by homozygous mutations in genes regulating cytotoxic and phagocytic immune cells, their role in the pathogenesis of IEC-HS is unknown. Here, we investigated the potential contribution of inheritable germline genetic variants to the risk of developing IEC-HS in adult lymphoma patients (pts) after CAR T therapy. Methods: We performed whole-exome sequencing on peripheral blood mononuclear cells (PBMCs) from 11 pts with IEC-HS (as per ASTCT diagnostic criteria) and 41 control lymphoma pts who did not develop IEC-HS after standard-of-care anti-CD19 CAR T therapy. The frequency of protein-truncating variants (PTVs) and pathogenic variants (as per ClinVar) in 587 genes associated with inherited disorders of the immune system including 51 genes implicated in primary HLH was analyzed. The effect on CAR T cells from alteration of UNC13D, the most frequently mutated gene in the IEC-HS cases, was studied in in vitro and in vivo lymphoma models. Results: When the 587 immune genes were analyzed, the number of PTVs and pathogenic variants was significantly more common in the cases vs. the controls (median 14/pt vs. 4/pt; p=3.6e-06). Analysis of the 51 primary HLH-implicated genes showed that the proportion of IEC-HS cases (7/11; 64%) with PTVs and pathogenic variants was considerably higher compared to the proportion of controls (11/41; 27%) (p<0.05). The IEC-HS cases had nonsense mutations in UNC13D and HAVCR2, frameshift mutations causing premature truncation in UNC13D, CYBC1, STXBP2, WAS, LAMP1, NLRP3, ADA and CDC42, and a pathogenic missense variant in TNFRSF1A; none of these genes were altered in the controls. All variants in the cases were heterozygous, and each gene was altered in a single case except for UNC13D and CYBC1, which were each mutated in two pts. Defects in these genes may be associated with excessive cytotoxic T cell activation due to impaired cytotoxic granule release (UNC13D) or excessive pro-inflammatory cytokine production by phagocytes (CYBC1), both hallmarks of HLH. To evaluate the impact of UNC13D alterations on CAR T cell function, we performed CRISPR/Cas9 knock-out (KO) of UNC13D in CAR T cells generated from healthy donor PBMCs. In co-culture assays, anti-CD19 CAR T cells with UNC13DKO were significantly less potent, had delayed cytotoxicity, and proliferated slower against B-cell lymphoma tumor cells compared with unmodified CAR T cells (p<0.0001). Consistent with the known function of UNC13D, degranulation was lower in UNC13DKOCD4 and CD8 CAR T cells. In contrast, production of pro-inflammatory cytokines such as IFN-γ, GM-CSF, and TNF-α was markedly higher in co-cultures with UNC13DKOCAR T cells (p<0.0001). In an in vivo B-cell lymphoma xenograft model, tumor clearance was delayed in mice treated with UNC13DKOanti-CD19 CAR T cells compared with unmodified CAR T cells. This was associated with a significant and sustained increase in pro-inflammatory cytokines (IFN-γ, GM-CSF, and TNF-α) in the serum of mice treated with UNC13DKOCAR T cells compared with unmodified CAR T cells (p< 0.0001). Evaluation of the effects of other germline variants on CAR T cells is ongoing. Conclusions: Our results suggest that rare inheritable germline variants in genes implicated in primary HLH may increase the risk of developing IEC-HS after CAR T cell therapy in adult lymphoma pts. While homozygous inheritance of these variants leads to primary HLH in childhood, the heterozygous inheritance observed in our pts likely contributed to the delayed onset of HLH in adulthood, triggered by CAR T cell therapy. Our studies with UNC13DKO CAR T cells reveal a plausible mechanism for IEC-HS development, involving excessive production of IFN-γ and other pro-inflammatory cytokines in these pts. Collectively, these findings enhance our understanding of the genetic and molecular mechanisms underlying IEC-HS, potentially aiding in the identification of pts at increased risk and informing the development of improved prophylactic and therapeutic interventions for this life-threatening complication.